The empowerment of women is a high-priority objective on international development agendas. It is also increasingly considered to be an important - if not fundamental - means of improving human welfare as well. In particular, the advancement of women is thought to increase investments in children. A small but growing body of research suggests that women value child welfare and the provision of public goods relatively more than men. However, development policy debate on this topic continues in part because the empirical literature on how women's status influences child health remains thin. The proposed research will provide new evidence on how women's suffrage influences infant and child mortality by drawing lessons from the historical United States. Early in the Twentieth Century, America witnessed both large advances in the status of women and unprecedented improvements in infant and child survival. These experiences present unrealized opportunities to learn about the relationship between the two and to advance the scientific understanding of historical mortality declines. This project has three specific aims: (1) To compile and disseminate annual state-level historical mortality statistics by cause and by age and sex in digital format (previously this data was only available in print), (2) To estimate how and through what pathways the expression of women's preferences through the electoral system influenced infant and child mortality, and (3) To examine whether protective public health and health education measures promoted by women are substitutes or complements for other major public health interventions such as drinking water disinfection (this is a critical but poorly understood determinant of intervention effectiveness). There are several attractive features of this research. First, early Twentieth Century America was remarkably similar to many developing countries today both in level of economic development and in epidemiological profile. It therefore provides a good natural laboratory for examining the effects of various development policies presently being discussed. Second, this project will take advantage of enormous variation in the timing of state-level women's suffrage laws created by America's system of federalism. Third, a variety of falsification exercises are available to test the validity of the empirical strategy. Finally, data on America during the early Twentieth Century is much richer than most data from developing countries today where comparable policies are being enacted. By developing and disseminating newly digitized historical state-level mortality data, this project will aid other researchers working in this area as well. [unreadable] [unreadable] [unreadable]